TWI394141B - Karaoke song accompaniment automatic scoring method - Google Patents

Description

Karaoke song accompaniment automatic scoring method

The invention relates to an automatic scoring method for karaoke song accompaniment, in particular to an innovative designer who calculates scores by weighted scoring method based on multiple scores such as sensation of sound, rhythm and emotion.

Press, in the karaoke (KARAOK) song accompaniment process, the current phonograph is usually accompanied by automatic scoring function, but the conventional design of this function is often only a rough estimate of the overall score, or may be based solely on the singing voice The decibel value is the only reference for the evaluation, and the scores of some phonographs are not related to the quality of the songs. Therefore, only a little entertainment effect can be achieved, and it cannot be truly judged. The songs are good or bad, so it doesn't really help the singer's practice.

Therefore, in view of the problems existing in the design and use of the above-mentioned karaoke song singer products, how to develop an innovative design that can be more ideal and practical, and the relevant industry should further consider the goal and direction of breakthrough.

In view of this, the inventor has been engaged in the manufacturing development and design experience of related products for many years. After detailed design and careful evaluation, the inventor has finally obtained the practical invention.

The main object of the present invention is to provide an automatic scoring method for karaoke song accompaniment, and the problem to be solved is that the automatic scoring function of the conventional karaoke song phonograph can not really judge the singing quality, so that The singer does not have the problem of helping to sing and think about the breakthrough. The technical feature of the problem solving of the present invention lies in the automatic scoring method of the karaoke song accompaniment, mainly by comparing the pitch and beat of the singer. The position and volume and the pitch, beat position and volume of the main melody of the song are respectively obtained into the pitch score, the rhythm score and the emotional score, and finally the weighted total score is calculated by weighted scoring method; thereby, the innovative unique design makes the present invention compare the previous In terms of technology, the pitch, beat position and volume error of the singer in each song passage can be accurately calculated, and the display effect of the pitch curve and the volume curve can be utilized, so that the singer can easily know which place sings. It is not accurate enough and where it needs to be strengthened. It has the dual effects of teaching and entertainment and is practical and progressive.

Please refer to the first embodiment of the present invention, which is a preferred embodiment of the automatic karaoke song singer scoring method of the present invention, but the embodiments are for illustrative purposes only, and are not limited by the structure in the patent application; The karaoke song accompaniment automatic scoring method is generally obtained by comparing the pitch of the singer, the position and volume of the beat with the pitch of the main melody of the song, the position of the beat and the volume, respectively. The scoring items of the rhythm score and the emotion score are finally calculated by weighted scoring method to calculate the weighted total score of the scoring items to obtain the score of the automatic scoring.

When a person is singing, in addition to the characteristics of personal voice, to comment The matching of songs and songs should mainly include three kinds of feelings, one is the sense of sound, the other is the sense of rhythm, the third is emotion, the sense of sound is the pitch accuracy of each pitch and the relative note; the sense of rhythm is to judge its beat The error of the point position includes the vocal beat point and the end beat point; the emotion is to judge the change of the volume, including the volume change of each sentence and the overall volume change. The methods for specifically obtaining the pitch score, the rhythm score, and the emotion score are respectively described as follows:

(1) Sound score:

Please refer to Figure 1 for a short period of time (for example, 0.1 seconds). The pitch of the singer is calculated by the microphone audio sung by the singer. This pitch estimation is the fundamental frequency of the person's voice. Fundamental Frequency), and its acquisition method can usually be obtained by the method based on Autocorrelation Function. Then, the fundamental frequency is first converted into a relative scale by the sound estimator, and then the sound level and the music main melody are compared. The degree of matching of the scales captured in the middle, and giving the scale a pitch score, thus calculating the pitch scores of all scales, until the end of the concert, an average pitch score can be output. As shown in Fig. 2, the details are as follows: First, "initial parameter setting", in which the number of scales initialized is m = 0, and the treble and the pitch of the scale are matched with NoteHit = 0, and the bass sense is matched. NoteHitAround=0, NoteHit indicates the number of time periods during which the human voice is highly matched during the performance of the scale, and NoteHitAround indicates the number of time periods when the human voice is higher than the musical scale within one semitone, and then the main melody of the next period is obtained. The scale and the calculation of the sound of a person with a high sound for a period of time, and the calculation of the pitch score algorithm of the scale are determined by the high-pitched sound matching value NoteHit, the low-pitched sound level matching value NoteHitAround and the scale length NoteLength; the main melody scale is directly obtained from the midi and other documents. Obtaining the scale of the performance relative to the time according to the increase of time, the human voice is high (the fundamental frequency), and the scale relative to the pitch can be converted through the transcoding table, for example, the frequency of the scale "A4" is 440 Hz, each Raise the octave, the frequency is increased by two times, such as the frequency of the scale "A5" is 880 Hz, an octave has 12 semitones, the frequency phase between each semitone 2 ^(1/12) times, because if the difference between the frequency of the vocal scale twice or 1/2 times the like, when an integer multiple relationship, the sense of sound is the same, thus the scale through ± 12 semitones, we adjusted the calculation The obtained human sound level Note_p and the scale of the main melody Note _{_m} make it differ between +6 semitones and -5 semitones, that is, Note_p=Note_p+12*i, i is an integer other than 0, so that -5<= Note_p-Note_m<=6. Next, determine whether it is a new scale, if yes, calculate the pitch score of the previous scale, and then reset the starting parameters, NoteHit=0 and NoteHitAround=0 and the number of scales m=m+1, if otherwise, compare the main melody scale with Human sound level matching, this matching means that the error is within a relatively small allowable range, such as 0.5 semitones. If it matches, the high-pitched sound matching value of the scale is increased NoteHit=NoteHit+1, otherwise it is judged whether the main melody scale Matches the human sound level to the bass sense. This bass sense match indicates that the error is within a relatively large allowable range, such as within one semitone, if so, increase the scale bass match value NoteHitAround=NoteHitAround+1, and then return to get The main melody scale and the calculation of the human voice are high for a while. The above “calculating the pitch score of the previous scale”, the algorithm is as shown in Fig. 3: first obtain the previous music main string scale length NoteLength(m), where: m=0, 1, 2, ..., M

The M is the total number of scales, and then it is judged whether the high-pitched sound matching value NoteHit is greater than zero, and if so, the high-pitched scale matching score is calculated: PitchScore(m)=PSH+K1 * NoteHit(m)/NoteLength(m); where: PSH, K1 is an adjustable empirical value parameter, otherwise calculate the bass level scale matching score: PitchScore(m)=PSL+K2 * NoteHitAround (m)/NoteLength(m); where: PSL, K2 are adjustable empirical value parameters, and limited: 0<=PitchScore(m)<=100

Finally, it is judged whether it is the last scale. If the above process is repeated, if it is, then “calculate the average pitch score”, the algorithm is the weighted average of all PitchScore(m) with the weighted NoteLength(m) as the weighted proportion, as follows: Degree NL=Σ _m=0~M-1 NoteLength(m), average score SOP(Score of Pitch):

(2) Rhythm score:

The sense of rhythm is determined by calculating the degree of matching between the vocal beat point and the start time of the music main melody scale and the ending time of the vocal end beat point and the end time of the music main melody scale. To accurately estimate the position of the singer's beat at each beat, here we estimate the singer's pitch as a change in the time of the different scales, and then determine the accuracy of the beat, as shown in Figure 4. It is shown that the method described in Figure 1 first estimates the pitch of the human voice and The scale of the main melody of the music is obtained, and then the average rhythm score is generated by the rhythm estimator.

Through the rhythm estimator, the human voice is first converted to a relative scale, and then the time is compared with the scale obtained from the main melody. The error of this time includes the early or delayed attack and end. The points are taken, and the time error of each scale is recorded, and then the rhythm score of the scale is given, so that the rhythm scores of all the scales are calculated until the end of the singing, and then the average rhythm score is output. As shown in Fig. 5, the rhythm delay matching device and the rhythm sense lead matcher can be used to calculate the vocal and the scale from the converted human sound level, the current, the previous and the next music main chord scale, respectively. The degree of delay in the time delay or advancement, the delay time and the lead time of the vocal end beat or the vocal beat, and then the rhythm score of the scale is obtained by calculating the rhythm sense score, according to which, from the first At the beginning of the scale, we calculate the rhythm error for each scale until the end of the last scale, and then calculate the average rhythm score.

Please refer to Figure 6, the rhythm-sensing delay matcher first determines whether it is the beginning of a new music scale, otherwise it determines whether the slap beat delay time has been set, and if so, it ends, otherwise it judges the human sound level and Whether the music scale matches, if otherwise increase the slap beat delay time, if yes, set the slap beat delay time, and then end, this delay time indicates the time error of the vocal start later than the music scale starts; if it is new music At the beginning of the scale, the singer beat delay time is reset and the last scale end time is recorded, and then it is judged whether the human sound level matches the previous music main chord scale, and if so, whether the next personal sound level is the same as the previous music master. The scale scale matches until no, then sets the end beat delay time after the knot Bundle, this delay time indicates the time error of the vocal ending later than the end of the last musical scale.

Please refer to Figure 7, the rhythm sense lead matcher, first determine whether it is the beginning of the new music scale, if not, determine whether the human sound level matches the current music scale, and if so, record the human sound level End time, otherwise set the end beat time, then end, this lead time indicates the time error before the end of the music scale, the vocal is earlier than it ends; if it is the beginning of the new music scale, reset the end beat time and Recording the scale start time, and then determining whether the human sound level matches the music main string scale, and if so, determining whether the previous human sound level matches the scale, until no, and then setting the start beat time to end, This lead time represents the time error that the vocal sound started earlier than it started before the musical scale began.

Then, the vocal beat delay time, the slap beat lead time, the end beat delay time, and the end beat lead time are calculated, and the scale beat score SOB (Score of Beat) is calculated. The algorithm is as follows: The error is TDS, then the vocal beat score (SOBS): SOBS = As +100. (1- TDS/Ls )

Among them, TDS = singer beat delay time (NoteOnLag) + singer beat point lead time (NoteOnLead), As and Ls are preset experience value parameters. Let the end time error be TDE, then: end score (SOBE): SOBE = Ae +100. (1- TDE/Le )

Among them, TDE=End beat delay time (NoteOffLag)+End beat time (NoteOffLead), Ae and Le are preset experience value parameters, the scale rhythm score (SOB): SOB = SOBS . R + SOBE . (1- R )

Where R is a preset weighting parameter and 0<=R<=1.

(3) Emotional scores:

Emotion is a parameter that is difficult to measure objectively. Here we use the degree of matching between the average amplitude of the calculated human voice and the average amplitude of the main music melody. Calculating the emotional score firstly obtains and calculates the degree of matching of the whole song and each sentence of the vocal amplitude curve and the music amplitude curve and the degree of change of the amplitude curve of each sentence relative to the overall amplitude to obtain an average emotional score. The average amplitude of the vocal is obtained by calculating the RMS (Root of Mean Square, hereinafter referred to as RMS) value of each individual voice segment. The average amplitude of the music main melody can also be calculated by calculating the RMS value of each main melody voice segment or directly. Obtained from the amplitude parameter in the synthesized music information, the algorithm of the RMS is as follows:

Where x(i), i=0,1,...,K-1,K represents the sound sample points (Samples) of this sound segment, and the RMS value can be used in other methods such as averaging. Replace with methods such as amplitude or maximum amplitude. As shown in Figures 8 and 9, the RMS value of the vocal signal and the music main melody is calculated by the emotion score estimator at intervals (about 0.1 sec), and the RMS sequence of the vocal and music can be obtained, assuming MicVol respectively. (n) and MelVol(n), n=0, 1, N-1, ,, represent the RMS value of the nth time period, where N is the total length of the song time and the energy of MicVol(n) The level is adjusted to be the same as MelVol(n), and then averaged according to the length of each scale, and the average RMS sequence of the mth scale of vocals and music is AvgMelVol(m) and AvgMicVol(m), respectively; It should be explained that the above MicVol represents the microphone voice volume (the same applies below), MelVol represents the music's main melody volume (the same below), and AvgMelVol (m) represents the average value of the music in the mth scale, AvgMicVol(m) It represents the average value of the microphone human voice in the mth scale; the scale represents the pitch of the note, and the length of the scale refers to the length of the note; assuming that the vocal amplitude and the music main melody amplitude sequence of the nth time period are respectively MicVol(n) and MelVol(n), n=0, 1, N-1, where N is the total length of the song time It is averaged according to the length of each scale, and the average amplitude sequence of the mth scale of vocals and music is AvgMelVol(m) and AvgMicVol(m), respectively;

Where L _m is the length of the mth note and n _m is the time period from the beginning of the mth note.

AvgMelVol(m), AvgMicVol(m) can be used to calculate the score of SOE (Score of Emotion). Firstly, the overall matching degree SOET of the vocal amplitude curve and the music amplitude curve is obtained and calculated, which can represent the overall emotional change score, as follows:

Where M is the total number of scales, and

Therefore SOET<=100.

Then, the calculation of each emotional score SOES can be performed. First, AvgMicVol(m) and AvgMelVol(m) are cut into one sentence, assuming that the starting scale of each lyric is S(j), j=0, 1, 2,..., L-1 starts, where L is the total number of lyrics, and let S(L)=M, then the emotional change score of each sentence is:

j=0,1,2,...,L-1, then calculate the relative emotional change score of each sentence, which is the change of the volume of each sentence relative to the overall volume: first, order

then

From the above, the average sentiment score is the weighted average of the overall emotional change score, the emotional change score of each sentence, and the relative emotional change score of each sentence:

Where α, β, and γ are weighting coefficients, and α+β+γ=1.

(4) Weighted total score: (Please refer to Figure 9)

The AES (Average Evaluated Score) obtained by the above SOP, SOB, and SOE is as follows: AES = p . SOP + q . SOB + r . SOE

Where p, q, r are weighting coefficients, and p+q+r=1.

Example of implementation:

Taking a song as an example, we calculate the vocal pitch MicPitch(n) and the RMS mean MicVol(n) every 0.1 seconds, while taking the pitch of the music main melody note MelNote(n) and calculating its RMS average MelVol (n), n=0,1,2,...,N,N represents the total length of the song, and there is no loss of generality. For convenience of explanation, N=280 is used here, indicating that the total length of the song is 28 seconds, such as Figure 10 is a graph of MicPitch(n) and MelNote(n). The solid line in the figure represents the pitch of the main melody note, and the vertical axis represents the music and human sound level. Each integer interval is a semitone, 60 means The midrange Do, 61 means the midrange rises Do, 69 means the midrange La, and so on, the dot represents the pitch calculated by the human voice, and turns it into a scale code, which has passed the positive and negative 12 semitones. Adjustment, so that the human voice is closest to the pitch of the main melody note, The line is a segment, each segment represents a continuous scale, the height of each segment is undulating, indicating the change of the scale. When the main melody scale is -1, indicating that the note is a rest or an empty scale, the ignore will be skipped. When the dot is zero, it means that the vocal is not calculated, and the vocal may be silent, mute or murmur, etc., and will be regarded as unvoiced.

First, by the algorithm of the above-mentioned pitch score, the high-pitched sound matching value NoteHit(m) of the mth scale (as shown by the circle in Fig. 11) and the bass sense matching value NoteHitAround(m) can be obtained (as in Fig. 11). As shown by the triangle), m=0,1,2,...M, M=3, as shown in Fig. 11, let PSH=50, K1=100, and PSL=35, K2=50, each scale can be obtained. The pitch score of m (as indicated by the rectangle in Fig. 11) is calculated by the weighted average of the note length (as indicated by the star in Fig. 11) to obtain an average pitch score ScoreOfPitch (SOP) = 98.

Then, by the above algorithm of the rhythm score, as shown in Fig. 12, NoteOnLag(m) (circle) and NoteOnLead(m) (star) of the mth scale can be obtained, so that As=10, Ls=10, You can calculate BeatOnScore(m) (rectangle), as shown in Figure 13, you can get NoteOffLag(m) (circle) and NoteOffLead(m) (star), let Ae=50, Le=NoteLength (scale length), BeatOffScore(m) (circle) can be calculated. After weighted average calculation of scale length, ScoreOfBeatStart(SOBS)=93.19, ScoreOfBeatEnd(SOBE)=99.82, let R=0.5, SOB=96.5.

Then, by the algorithm of the above emotion score, the RMS sequence MelVol(n) of the vocal and music main melody (as shown by L1 in Fig. 14) and MicVol(n) (as shown by L2 in Fig. 14) can be obtained first. And adjust the energy level of MicVol(n) to be the same as MelVol(n). As shown in Figure 14, it is based on each scale. The average length of the mth scale can be obtained by the average RMS sequence of the mth scale, AvgMelVol(m) (as shown in Fig. 15L3), and AvgMicVol(m) (as shown in Fig. 15L4), as shown in Fig. 15, setting Weighting coefficient, and thus SOET=98.33, SOES(j) of the jth sentence (as shown in Fig. 16L5) and SOEA(j) (as shown in Fig. 16L6), j=0,1, 2, ... L-1, the total number of sentences L = 6, as shown in Figure 16, the average SOES = 97.2, and SOEA = 95.67, after weighting calculations: ScoreOfEmotion (SOE) = 97.24

Finally, the weighting coefficient p=0.6, q=0.2, and r=0.2 are set, and the weighted total score can be obtained: AES = p . SOP + q . SOB + r . SOE =97.55

Advantages of the invention:

The karaoke song accompaniment automatic scoring method of the present invention mainly obtains the pitch score, the rhythm score and the emotion score by comparing the pitch of the singer, the position and volume of the beat, the pitch of the main melody of the song, the position of the beat and the volume. The innovative unique design of the weighted total score is calculated by weighted scoring method, so that the present invention can accurately calculate the pitch, the beat position and the volume error of the singer in each song passage according to the prior art, and can utilize The display of pitch curve and volume curve makes it easy for the singer to know where the sing is not accurate enough and where it needs to be strengthened, to achieve practical progress with both teaching and entertainment effects.

The present invention has been specifically described by the above embodiments, and although the description is by specific terms, the invention cannot be limited thereto. The scope of the invention is subject to change and modification of the spirit and principles of the present invention, and such changes and modifications are intended to be included as described below. In the scope defined by the scope of application for patents.

Fig. 1 is a block diagram 1 of a method for obtaining a pitch score of the present invention.

Fig. 2 is a block diagram 2 of the method for obtaining the pitch of the sound of the present invention.

Fig. 3 is a block diagram 3 of the method for obtaining the pitch score of the present invention.

Fig. 4 is a block diagram 1 of the method for obtaining the rhythm score of the present invention.

Fig. 5 is a block diagram 2 of the method for obtaining the rhythm score of the present invention.

Fig. 6 is a block diagram 3 of the method for obtaining the rhythm score of the present invention.

Fig. 7 is a block diagram 4 of the method for obtaining the rhythm score of the present invention.

Figure 8 is a block diagram of the method for obtaining the sentiment score of the present invention.

Figure 9 is a block diagram of the automatic score estimation method of the present invention.

Figure 10: An example of the implementation of the present invention is illustrated in Figure 1.

Figure 11: An example of the implementation of the present invention is illustrated in Figure 2.

Figure 12: An example of the implementation of the present invention is illustrated in Figure 3.

Figure 13: An example of the implementation of the present invention is illustrated in Figure 4.

Figure 14: A practical example of the invention is illustrated in Figure 5.

Figure 15: An example of the implementation of the present invention is illustrated in Figure 6.

Figure 16: An example of the implementation of the present invention is illustrated in Figure 7.

Claims (2)

A method for automatically scoring a karaoke song singer, mainly by comparing the pitch of the singer, the position and volume of the beat with the pitch of the main melody of the music, the position of the beat and the volume, to obtain the score of the sensation and the score of the rhythm, respectively. The scoring items of the emotional scores are finally calculated by weighted scoring method to calculate the weighted total scores of the scoring items to obtain the scores of the automatic scoring; wherein the pitch scores are obtained by the singer every other short period of time The sing microphone audio estimates the pitch of a singer. The estimation of the pitch is obtained by the fundamental frequency of the human voice, and then the fundamental frequency is first converted into a relative scale by a sound estimator, and then Comparing the degree of matching between the person's sound level and the scale captured in the main theme of the music, and giving the scale a pitch score, thus calculating the pitch score of all scales, until the end of the concert, an average pitch score can be output; The score is obtained by including the initial parameter setting, setting the number of scales that are initialized, m=0, and the matching value of the vocal and the treble of the scale. eHit=0, and the bass sense match value NoteHitAround=0, NoteHit indicates the number of time periods when the human voice is highly matched during the scale playing, and NoteHitAround indicates the number of time periods when the human voice is high and the scale is within one semitone, and then Obtaining the main melody scale for the next period of time and calculating the human voice for a period of time; and calculating the pitch score algorithm of the scale, which is determined by the high-pitched sound matching value NoteHit, the low-pitched sound matching value NoteHitAround, and the scale length NoteLength; The rhythm sense score is determined by calculating the vocal beat point and the start time of the music main melody scale and the matching degree of the vocal end beat point and the end time of the music main melody scale; the person is first introduced by the rhythm estimator The sound is converted to a relative scale and then compared to the time scale of the scale obtained from the scale and the main melody. The error of this time includes the early or delayed start and end beats, and the time of each scale is recorded. Error, then give the rhythm sense score of the scale, thus calculate the rhythm score of all scales until the end of the concert, and then output the average rhythm score; use the rhythm sense delay matcher and the rhythm sense lead matcher, after conversion The human sound level, the current, the previous and the next music main string scale, respectively calculate the degree of matching between the human voice and the scale in time delay or advance, and obtain the delay time and the lead time of the vocal end beat or the vocal beat. Then, by calculating the scale rhythm score, the rhythm score of the scale is obtained, and accordingly, starting from the first scale The rhythm of each scale is errored until the last scale ends, and then the average rhythm score is calculated; wherein the sentiment score is determined by the degree to which the average amplitude of the calculated human voice matches the average amplitude of the musical main melody; The average amplitude of the vocal is obtained by calculating the RMS (Root of Mean Square) value of each individual voice section, and the average amplitude of the music main melody is calculated by calculating the RMS value of each main melody sound section or directly by the synthesized music. The amplitude parameter in the information is obtained; the RMS value of the vocal signal and the music main melody is calculated by the emotion score estimator at intervals, to obtain the RMS sequence of the vocal and music, and then averaged according to the length of each scale. , the sound of vocals and music Order average RMS sequence; then calculate the emotional score, first obtain and calculate the degree of matching between the whole song and each sentence of the vocal amplitude curve and the music amplitude curve and the degree of change of the amplitude curve of each sentence relative to the overall amplitude to obtain an average emotion fraction. According to the automatic karaoke song accompaniment scoring method described in claim 1, wherein if the high-pitched sound matching value NoteHit is greater than zero, the m-th high-pitched scale matching score is calculated (abbreviated as PitchScore(m)): PitchScore(m) =PSH+K1 * NoteHit(m)/NoteLength(m) where: PSH, K1 are adjustable empirical value parameters, otherwise calculate the bass level scale matching score: PitchScore(m)=PSL+K2 * NoteHitAround(m)/NoteLength (m) where: PSL, K2 are adjustable empirical value parameters, and limit: 0 <= PitchScore (m) <= 100 to determine whether it is the last scale, otherwise repeat the above process, if yes then "calculate the average pitch score The algorithm is a weighted average of all PitchScore(m) weighted proportions of NoteLength(m) as follows: Let the total length of the scale NL = Σ _{m = 0~M-1} NoteLength(m), the average pitch score SOP ( Score of Pitch): The algorithm for calculating the score of the score of SOB (Score of Beat) is as follows: the vocal beat time error is TDS, and the sing beat score (SOBS): SOBS = As +100. (1- TDS/Ls ) where TDS = singer beat delay time (NoteOnLag) + singer beat lead time (NoteOnLead), As and Ls are preset experience value parameters, so that the end beat time error is TDE , then: end beat score (SOBE): SOBE = Ae +100. (1- TDE/Le ) where TDE = end beat delay time (NoteOffLag) + end beat lead time (NoteOffLead), Ae and Le are preset experience value parameters, the scale rhythm sense score (SOB): SOB = SOBS . R + SOBE . (1- R ) where R is a preset weighting parameter, and 0<=R<=1 assumes that the vocal amplitude and the music main melody amplitude sequence of the nth time period are MicVol(n) and MelVol(n), respectively. n=0, 1, N-1, where N is the total length of the song, which is averaged according to the length of each scale, and the average amplitude sequence of the mth scale of the vocals and music is AvgMelVol(m), AvgMicVol, respectively. (m); as follows: Where L _m is the length of the mth note, n _m is the time period from the beginning of the mth note; AvgMelVol(m), AvgMicVol(m) is used to calculate the score of SOE (Score of Emotion), first obtained and calculated The overall matching degree of the vocal amplitude curve and the music amplitude curve, SOET, which represents the overall emotional change score, as follows: Where M is the total number of scales and: Therefore, SOET<=100 is followed by the calculation of each sentiment score SOES. First, the AvgMicVol(m), AvgMelVol(m) is cut into one sentence, assuming that the starting scale of each lyric is S(j), j =0,1,2,..., L-1 start, where L is the total number of lyrics, and let S(L)=M, then the emotional change score of each sentence is: j=0,1,2,...,L-1, then calculate the relative emotional change score of each sentence, which is the change of the volume of each sentence relative to the overall volume: first, let: then The average sentiment score obtained from the above is the weighted average of the overall emotional change score, the emotional change score of each sentence, and the relative emotional change score of each sentence: Where α, β, γ are weighting coefficients, and α+β+γ=1; the AES (Average Evaluated Score) obtained by the SOP, SOB, and SOE is as follows: AES = p . SOP + q . SOB + r . SOE where p, q, r are weighting coefficients, and p+q+r=1.